You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Milind Bhandarkar (JIRA)" <ji...@apache.org> on 2007/03/06 19:05:24 UTC
[jira] Created: (HADOOP-1069) Rename Hadoop record I/O to Jute
Rename Hadoop record I/O to Jute
--------------------------------
Key: HADOOP-1069
URL: https://issues.apache.org/jira/browse/HADOOP-1069
Project: Hadoop
Issue Type: Improvement
Components: record
Affects Versions: 0.12.0
Environment: All
Reporter: Milind Bhandarkar
Assigned To: Milind Bhandarkar
Fix For: 0.13.0
jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "Doug Cutting (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12478861 ]
Doug Cutting commented on HADOOP-1069:
--------------------------------------
> Our goal is to assign "Jute" a name that is short, memorable and works with our goals of defining a top-level package that can be used in many contexts.
The record package is currently a part of Hadoop. If it moves out of Hadoop then it will require a different name, outside of the org.apache.hadoop package. If and when that ever happens, then the package name would probably be changed to match the project name, just as many org.apache.nutch packages were renamed to be org.apache.hadoop when Hadoop was split from Nutch. Until then, I see no problems with the package name org.apache.hadoop.record that warrant changing it. It is in accord with our naming guidelines and org.apache.hadoop.jute is not: all but project names are descriptive.
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Assigned To: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "eric baldeschwieler (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12478857 ]
eric baldeschwieler commented on HADOOP-1069:
---------------------------------------------
So let's get some proposals for good names?
Our goal is to assign "Jute" a name that is short, memorable and works with our goals of defining a top-level package that can be used in many contexts. Doesn't seem optimal. "Jute" seems to do a better job.
Open to other suggestions.
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Assigned To: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "Doug Cutting (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12478498 ]
Doug Cutting commented on HADOOP-1069:
--------------------------------------
-1 The general naming rule is that projects should have meaningless names but that everything else should strive to have a descriptive name. Look for example at Java itself: the term "Java" had little to do with computing, while the package names listed in http://java.sun.com/j2se/1.5.0/docs/api/overview-summary.html are descriptive.
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Assigned To: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "Owen O'Malley (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Owen O'Malley resolved HADOOP-1069.
-----------------------------------
Resolution: Won't Fix
Assignee: (was: Milind Bhandarkar)
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "Sameer Paranjpye (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12478516 ]
Sameer Paranjpye commented on HADOOP-1069:
------------------------------------------
What about 'beans' and 'swing'?
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Assigned To: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
Re: [jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by David Bowen <db...@yahoo-inc.com>.
>
> Swing was originally a separate project that was grafted into the JVM and not renamed, probably for back-compatibility.
>
FWIW I was in the Java group at Sun at the time. Swing was always
intended to be a new windowing toolkit for the Java platform, offering a
higher level of abstraction than the pre-existing Abstract Windowing
Toolkit, java.awt, which was already a bad package name. It had to be
called something. There could have been various problems with using a
sensible package name such as "windows".
[jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "Doug Cutting (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12478526 ]
Doug Cutting commented on HADOOP-1069:
--------------------------------------
> What about 'beans' and 'swing'?
Beans is meant to be descriptive, via a pun: a unit of Java (coffee) is a bean.
Swing was originally a separate project that was grafted into the JVM and not renamed, probably for back-compatibility.
But these are good examples of bad Java package names. Thanks!
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Assigned To: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-1069) Rename Hadoop record I/O to Jute
Posted by "Sameer Paranjpye (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-1069?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12478598 ]
Sameer Paranjpye commented on HADOOP-1069:
------------------------------------------
These are not bad names, IMO. I didn't have to walk down the list to come up with these, it was the first thought that crossed my mind.
FWIW, MapReduce is not a particularly descriptive name. It is a label that has been made popular by the MR paper. If a person hasn't read the paper, hears about Hadoop being a distributed computing platform, and then starts browsing the javadoc - the name 'org.apache.hadoop.mapred' doesn't bring up an image of a cluster performing computation.
There are cases where you need descriptive names and cases where you need labels or brands. The latter are by far less common than the former, but they do come up. They are often merited by applications or frameworks, this is such a case.
> Rename Hadoop record I/O to Jute
> --------------------------------
>
> Key: HADOOP-1069
> URL: https://issues.apache.org/jira/browse/HADOOP-1069
> Project: Hadoop
> Issue Type: Improvement
> Components: record
> Affects Versions: 0.12.0
> Environment: All
> Reporter: Milind Bhandarkar
> Assigned To: Milind Bhandarkar
> Fix For: 0.13.0
>
>
> jute was the original name of the hadoop record i/o component. IMHO, it is easier to pronounce, easier to remember and has already stuck among its users. This renaming should be done while there isn't a large codebase using jute, otherwise it will be very difficult later. rcc will be renamed jrc (jute record compiler).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.